Pred_Data.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Pred_Data.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "rh06AkGBs3et"
      ]
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Prediction of the rent prices \n",
        "After the cleaning and data processing part now let's jump to the main objective of this project the rent prediction part. For that we will use many regression based model as we will present further after. \n",
        "\n",
        "\n",
        "## 1.   Feature engineering \n",
        "\n",
        "1.   Changing catergorical data to ordinal data\n",
        "2.   Standardization of the data\n",
        "\n",
        "\n",
        "## 2.   Applying regression predictive model\n",
        "\n",
        "### Models to be used for the prediction\n",
        "> * Lasso Regression\n",
        "> * Ridge Regression\n",
        "> * Kfold Regularization"
      ],
      "metadata": {
        "id": "BaVBHiG1YH1e"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "* First let's take a look on our cleaned dataset, in order to do it let's import our previous results"
      ],
      "metadata": {
        "id": "IJBLzEqXabiM"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rFB91QvIX7ET",
        "outputId": "c42b2af7-55b8-41de-a579-2e00bed856eb"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/gdrive\n"
          ]
        }
      ],
      "source": [
        "# Access files from Google Drive\n",
        "from google.colab import drive \n",
        "drive.mount('/content/gdrive')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Necessary imports\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.pylab as plb \n",
        "%matplotlib inline\n",
        "\n",
        "import statsmodels.api as sm\n",
        "from sklearn.model_selection import train_test_split, KFold\n",
        "from sklearn.linear_model import LinearRegression, Lasso, LassoCV, Ridge, RidgeCV\n",
        "from sklearn.preprocessing import StandardScaler, PolynomialFeatures\n",
        "from sklearn.metrics import r2_score, mean_squared_error\n",
        "\n",
        "sns.set(context='notebook', style='whitegrid', font_scale=1.2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RqIgBu7bcwwx",
        "outputId": "c502f1b6-179d-4fff-aa0b-fe4ff88d7cc1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Importing the data and data processing"
      ],
      "metadata": {
        "id": "KitEQbCyDzzz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sf = pd.read_csv(\"./gdrive/My Drive/Projecto_BD/sf_clean.csv\")"
      ],
      "metadata": {
        "id": "WQLYTvjRc8Qo"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Re-adapting hood_district names\n",
        "sf['hood_district'] = sf['hood_district'].astype(int)\n",
        "sf['hood_district'] = sf['hood_district'].astype(str)"
      ],
      "metadata": {
        "id": "AkgCSpm5dqBS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sf.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "BW3k7fgTn0YL",
        "outputId": "e9b57371-8760-4c39-89ca-9a09b4f5f407"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-b2ddc688-1465-403f-9d66-e151763d058b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>price</th>\n",
              "      <th>sqft</th>\n",
              "      <th>beds</th>\n",
              "      <th>bath</th>\n",
              "      <th>laundry</th>\n",
              "      <th>pets</th>\n",
              "      <th>housing_type</th>\n",
              "      <th>parking</th>\n",
              "      <th>hood_district</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3920</td>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>(a) in-unit</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3870</td>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>(a) in-unit</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5105</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>(a) in-unit</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4850</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>(a) in-unit</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4750</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>(a) in-unit</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b2ddc688-1465-403f-9d66-e151763d058b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-b2ddc688-1465-403f-9d66-e151763d058b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-b2ddc688-1465-403f-9d66-e151763d058b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   price    sqft  beds  ...  housing_type        parking hood_district\n",
              "0   3920   805.0   1.0  ...     (c) multi  (b) protected             6\n",
              "1   3870   805.0   1.0  ...     (c) multi  (b) protected             6\n",
              "2   5105  1388.0   2.0  ...     (c) multi  (b) protected             6\n",
              "3   4850  1388.0   2.0  ...     (c) multi  (b) protected             6\n",
              "4   4750  1388.0   2.0  ...     (c) multi  (b) protected             6\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Limiting the scope of our data by limiting the outliers we're gonna pick to rental offers under **6000$**"
      ],
      "metadata": {
        "id": "VnaX4ecLpSuM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sf = sf[sf['price'] < 6000]"
      ],
      "metadata": {
        "id": "E77_ct2derH6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Converting nominal to ordinal variables "
      ],
      "metadata": {
        "id": "1x-i7UM5oRE7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sf.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Y2oiy539napc",
        "outputId": "f05356c9-7c1b-4706-9ff9-50688677ba7f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 987 entries, 0 to 1087\n",
            "Data columns (total 9 columns):\n",
            " #   Column         Non-Null Count  Dtype  \n",
            "---  ------         --------------  -----  \n",
            " 0   price          987 non-null    int64  \n",
            " 1   sqft           987 non-null    float64\n",
            " 2   beds           987 non-null    float64\n",
            " 3   bath           987 non-null    float64\n",
            " 4   laundry        987 non-null    object \n",
            " 5   pets           987 non-null    object \n",
            " 6   housing_type   987 non-null    object \n",
            " 7   parking        987 non-null    object \n",
            " 8   hood_district  987 non-null    object \n",
            "dtypes: float64(3), int64(1), object(5)\n",
            "memory usage: 77.1+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* As we can see our data have for categorical values :\n",
        "\n",
        "1.   Laundry\n",
        "2.   Pets\n",
        "3.   Housing type\n",
        "4.   Parking\n",
        "5.   Hood District\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "7HgYba2Bob4Y"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's begin first with Laundry for that we will create a dictionnary where each **nominal** value will have its **ordinal** equivalent "
      ],
      "metadata": {
        "id": "1RttbrKspnFN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "laundry_dict = {'(a) in-unit': 2, \n",
        "               '(b) on-site': 1,\n",
        "               '(c) no laundry': 0}\n",
        "sf = sf.replace({'laundry': laundry_dict})"
      ],
      "metadata": {
        "id": "X5xxPRZ9okHJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sf.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "dQm8JYzqpS3l",
        "outputId": "0f96d3f8-7404-4bc4-fb4b-fdd6cb5745a5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-36b8d7d4-e873-418e-94fa-ab952fa76737\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>price</th>\n",
              "      <th>sqft</th>\n",
              "      <th>beds</th>\n",
              "      <th>bath</th>\n",
              "      <th>laundry</th>\n",
              "      <th>pets</th>\n",
              "      <th>housing_type</th>\n",
              "      <th>parking</th>\n",
              "      <th>hood_district</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3920</td>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3870</td>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5105</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4850</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4750</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>(a) both</td>\n",
              "      <td>(c) multi</td>\n",
              "      <td>(b) protected</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-36b8d7d4-e873-418e-94fa-ab952fa76737')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-36b8d7d4-e873-418e-94fa-ab952fa76737 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-36b8d7d4-e873-418e-94fa-ab952fa76737');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   price    sqft  beds  ...  housing_type        parking hood_district\n",
              "0   3920   805.0   1.0  ...     (c) multi  (b) protected             6\n",
              "1   3870   805.0   1.0  ...     (c) multi  (b) protected             6\n",
              "2   5105  1388.0   2.0  ...     (c) multi  (b) protected             6\n",
              "3   4850  1388.0   2.0  ...     (c) multi  (b) protected             6\n",
              "4   4750  1388.0   2.0  ...     (c) multi  (b) protected             6\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's do the same for the other categorical values"
      ],
      "metadata": {
        "id": "cA_w7x31pgbm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sf.parking.unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6u3jfVB9okEt",
        "outputId": "d13a335d-f96e-4b84-d18b-7e60d4471bed"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['(b) protected', '(d) no parking', '(c) off-street', '(a) valet'],\n",
              "      dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "parking_dict = {'(a) valet': 3, \n",
        "               '(b) protected': 2,\n",
        "               '(c) off-street': 1,\n",
        "               '(d) no parking': 0}\n",
        "\n",
        "sf = sf.replace({'parking': parking_dict})"
      ],
      "metadata": {
        "id": "B5Bv-W9TqJdX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Same goes for pets but here we're gonna simplify even more the transformation by having a binary value "
      ],
      "metadata": {
        "id": "e4BYeobVqNW5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sf.pets.unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tFb5g1tNqQvz",
        "outputId": "61bd81d8-0111-4cef-fd97-4174c34ca98f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['(a) both', '(d) no pets', '(c) cats', '(b) dogs'], dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pets_dict = {'(a) both': 1, '(b) dogs': 1,\n",
        "            '(c) cats': 1,'(d) no pets': 0}\n",
        "sf = sf.replace({'pets': pets_dict})"
      ],
      "metadata": {
        "id": "60KKP6ebqWOz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sf.housing_type.unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "el1qgrCoq_kM",
        "outputId": "e1bebbac-9424-41c6-ac67-89d0cb925759"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['(c) multi', '(b) double', '(a) single'], dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As we've seen in EDA phase the housing type don't give any more information about the quality and the rent price of the appartement so we're gonna drop it "
      ],
      "metadata": {
        "id": "LpXcRXr3rcrq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sf=sf.drop(['housing_type'],axis=1)"
      ],
      "metadata": {
        "id": "-AS3KAVQrs6G"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now let's face our last categorical value the **'neighborhood districts'**"
      ],
      "metadata": {
        "id": "dFArZIQZr1Yl"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sf=pd.get_dummies(sf)"
      ],
      "metadata": {
        "id": "EWAFGRKcr_tI"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sf=sf.drop(['hood_district_10'],axis=1)"
      ],
      "metadata": {
        "id": "XGIRlpjosqdu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Encapsulate our Data into a csv file\n",
        "sf.to_csv('sf_final.csv', index=False)"
      ],
      "metadata": {
        "id": "o3cTSo-xaYr-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sf.head()"
      ],
      "metadata": {
        "id": "jpYUp8wEn-7T",
        "outputId": "6960288c-3cc4-40e7-8f32-7f3a1a709664",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 270
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-fbf289a2-8f2a-4f2f-a1af-8a8ebc7680bc\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>price</th>\n",
              "      <th>sqft</th>\n",
              "      <th>beds</th>\n",
              "      <th>bath</th>\n",
              "      <th>laundry</th>\n",
              "      <th>pets</th>\n",
              "      <th>parking</th>\n",
              "      <th>hood_district_1</th>\n",
              "      <th>hood_district_2</th>\n",
              "      <th>hood_district_3</th>\n",
              "      <th>hood_district_4</th>\n",
              "      <th>hood_district_5</th>\n",
              "      <th>hood_district_6</th>\n",
              "      <th>hood_district_7</th>\n",
              "      <th>hood_district_8</th>\n",
              "      <th>hood_district_9</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3920</td>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3870</td>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5105</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4850</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4750</td>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fbf289a2-8f2a-4f2f-a1af-8a8ebc7680bc')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-fbf289a2-8f2a-4f2f-a1af-8a8ebc7680bc button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-fbf289a2-8f2a-4f2f-a1af-8a8ebc7680bc');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   price    sqft  beds  ...  hood_district_7  hood_district_8  hood_district_9\n",
              "0   3920   805.0   1.0  ...                0                0                0\n",
              "1   3870   805.0   1.0  ...                0                0                0\n",
              "2   5105  1388.0   2.0  ...                0                0                0\n",
              "3   4850  1388.0   2.0  ...                0                0                0\n",
              "4   4750  1388.0   2.0  ...                0                0                0\n",
              "\n",
              "[5 rows x 16 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Modeling\n",
        "\n",
        "1. ***Train/Test split***"
      ],
      "metadata": {
        "id": "rh06AkGBs3et"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "* First let's get rid of our feature columns to predict which is the price here"
      ],
      "metadata": {
        "id": "Tjm2A4A0tJLL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X, y = sf.drop('price',axis=1), sf['price']\n",
        "\n",
        "#Train&Test\n",
        "# Let's choose 20% of the data for testing\n",
        "X, X_test, y, y_test = train_test_split(X, y, test_size=.2, random_state=42)\n",
        "\n",
        "# Training&Validation\n",
        "X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=.2, random_state=4)"
      ],
      "metadata": {
        "id": "GQHHs43PtFKg"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "poly = PolynomialFeatures(degree=2, interaction_only=False)\n",
        "    \n",
        "X_train_poly = poly.fit_transform(X_train)\n",
        "X_val_poly = poly.transform(X_val)\n",
        "X_test_poly = poly.transform(X_test)"
      ],
      "metadata": {
        "id": "8JoEniJJuLbA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* First in order to prepare our data to the regularization's techniques let's first apply standardization to avoid overfitting the data and lower their dispersion "
      ],
      "metadata": {
        "id": "MecLN6fJ8wyg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Scaling for Ridge / Lasso \n",
        "scaler = StandardScaler(with_mean=False)\n",
        "\n",
        "X_train_scaled = scaler.fit_transform(X_train_poly)\n",
        "X_val_scaled = scaler.transform(X_val_poly)\n",
        "X_test_scaled = scaler.transform(X_test_poly)"
      ],
      "metadata": {
        "id": "xppTqFyd8rPv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Machine learning models \n"
      ],
      "metadata": {
        "id": "e0exb38Lt_jn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 2. Regularization's techniques\n",
        "\n",
        "> **2.1 Lasso** :\n",
        "\n"
      ],
      "metadata": {
        "id": "FbHhsXSYvA2Z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "alphavec = 10**np.linspace(-2,1,10000)\n",
        "\n",
        "lasso_model = LassoCV(alphas=alphavec, cv=3).fit(X_train_scaled, y_train)\n"
      ],
      "metadata": {
        "id": "PpRMNj5LwNDg"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Let's identify the best alpha obtained from the cross validation "
      ],
      "metadata": {
        "id": "PPqTENOiwVKP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "lasso_alpha = lasso_model.alpha_\n",
        "lasso_alpha"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F80hzqrYwhTL",
        "outputId": "21d27f13-0ac9-4c9a-91ca-6adf5f28dcbf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.3731366893478814"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now that we have our best parameter let's apply the lasso's regressor to our training set  "
      ],
      "metadata": {
        "id": "IPxCtbpKxGJV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "las=Lasso(alpha=lasso_alpha)\n",
        "las.fit(X_train_scaled,y_train)\n",
        "#Using R^2 to compute the accuracy of the pred\n",
        "print(f'Lasso Regression train R^2: {las.score(X_train_scaled, y_train):.5f}')\n",
        "print(f'Lasso Regression val R^2: {las.score(X_val_scaled, y_val):.5f}')\n",
        "print(\"\")\n",
        "#Using RMSE to compute the error\n",
        "print(f'Lasso Regression RMSE (train): {mean_squared_error(y_train, las.predict(X_train_scaled), squared=False):.5f}')\n",
        "print(f'Lasso Regression RMSE (val): {mean_squared_error(y_val, las.predict(X_val_scaled), squared=False):.5f}')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w7a-jAiSxFbB",
        "outputId": "810fb5b1-1e46-46f2-9215-daf54643c5bb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Lasso Regression train R^2: 0.77259\n",
            "Lasso Regression val R^2: 0.59567\n",
            "\n",
            "Lasso Regression RMSE (train): 516.74670\n",
            "Lasso Regression RMSE (val): 674.18450\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* The score is high in training and average one in Validation for both **RMSE** and **R^2**"
      ],
      "metadata": {
        "id": "GGc_bgMfyHjt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "> **2.2 Ridge** :\n",
        "\n"
      ],
      "metadata": {
        "id": "VZg_0XCZyZEX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "alphavecr=10*np.linspace(-1,1,10000)\n",
        "\n",
        "ridge_model=RidgeCV(alphas=alphavecr,cv=3)\n",
        "ridge_model.fit(X_train_scaled,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fSQz38gUykpg",
        "outputId": "95399114-9963-4c8f-c666-e0f0b48fdd35"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RidgeCV(alphas=array([-10.       ,  -9.9979998,  -9.9959996, ...,   9.9959996,\n",
              "         9.9979998,  10.       ]),\n",
              "        cv=3)"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "ridge_alpha=ridge_model.alpha_\n",
        "ridge_alpha"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AMnpJZu0y_XF",
        "outputId": "d27436b7-844c-4349-d94b-06b9a6bae50d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.341134113411342"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rid=Ridge(alpha=ridge_alpha)\n",
        "rid.fit(X_train_scaled,y_train)\n",
        "print(f'Ridge Regression train R^2: {rid.score(X_train_scaled, y_train):.5f}')\n",
        "print(f'Ridge Regression val R^2: {rid.score(X_val_scaled, y_val):.5f}')\n",
        "print(\"\")\n",
        "print(f'Ridge Regression RMSE (train): {mean_squared_error(y_train, rid.predict(X_train_scaled), squared=False):.5f}')\n",
        "print(f'Ridge Regression RMSE (val): {mean_squared_error(y_val, rid.predict(X_val_scaled), squared=False):.5f}')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bxS7QrIOzMgg",
        "outputId": "7d6bb970-e4b5-4b7a-f456-ec6de1d324e0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Ridge Regression train R^2: 0.77580\n",
            "Ridge Regression val R^2: 0.60265\n",
            "\n",
            "Ridge Regression RMSE (train): 513.08999\n",
            "Ridge Regression RMSE (val): 668.33680\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* With Try with some other parameters for alpha to see if there's any amelioration"
      ],
      "metadata": {
        "id": "K-vbQCoX10bs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "alphavecr=10*np.linspace(-2,2,500)\n",
        "\n",
        "ridge_model=RidgeCV(alphas=alphavecr,cv=3)\n",
        "ridge_model.fit(X_train_scaled,y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yFLFtx8i1y5G",
        "outputId": "9eb03617-4716-43ba-abfc-7a4ec1a5ff82"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RidgeCV(alphas=array([-20.        , -19.91983968, -19.83967936, -19.75951904,\n",
              "       -19.67935872, -19.5991984 , -19.51903808, -19.43887776,\n",
              "       -19.35871743, -19.27855711, -19.19839679, -19.11823647,\n",
              "       -19.03807615, -18.95791583, -18.87775551, -18.79759519,\n",
              "       -18.71743487, -18.63727455, -18.55711423, -18.47695391,\n",
              "       -18.39679359, -18.31663327, -18.23647295, -18.15631263,\n",
              "       -18.0761523 , -17.99599198,...\n",
              "        17.51503006,  17.59519038,  17.6753507 ,  17.75551102,\n",
              "        17.83567134,  17.91583166,  17.99599198,  18.0761523 ,\n",
              "        18.15631263,  18.23647295,  18.31663327,  18.39679359,\n",
              "        18.47695391,  18.55711423,  18.63727455,  18.71743487,\n",
              "        18.79759519,  18.87775551,  18.95791583,  19.03807615,\n",
              "        19.11823647,  19.19839679,  19.27855711,  19.35871743,\n",
              "        19.43887776,  19.51903808,  19.5991984 ,  19.67935872,\n",
              "        19.75951904,  19.83967936,  19.91983968,  20.        ]),\n",
              "        cv=3)"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "ridge_alpha=ridge_model.alpha_\n",
        "ridge_alpha"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T_ONFooC1zkC",
        "outputId": "2b34ce36-f7ed-47f3-aa69-6a56f2699fff"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.3226452905811614"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rid=Ridge(alpha=ridge_alpha)\n",
        "rid.fit(X_train_scaled,y_train)\n",
        "print(f'Ridge Regression train R^2: {rid.score(X_train_scaled, y_train):.5f}')\n",
        "print(f'Ridge Regression val R^2: {rid.score(X_val_scaled, y_val):.5f}')\n",
        "print(\"\")\n",
        "print(f'Ridge Regression RMSE (train): {mean_squared_error(y_train, rid.predict(X_train_scaled), squared=False):.5f}')\n",
        "print(f'Ridge Regression RMSE (val): {mean_squared_error(y_val, rid.predict(X_val_scaled), squared=False):.5f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x9cDTSMH2GYT",
        "outputId": "6ae73dbd-7715-4f21-a702-23ce208bbf3a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Ridge Regression train R^2: 0.77582\n",
            "Ridge Regression val R^2: 0.60281\n",
            "\n",
            "Ridge Regression RMSE (train): 513.06055\n",
            "Ridge Regression RMSE (val): 668.20153\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Let's decide between both of the regressor the best predictor using KFold"
      ],
      "metadata": {
        "id": "MMxO6Se72NYR"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "> **2.3 Kfold for Ridge and Lasso**\n",
        "\n"
      ],
      "metadata": {
        "id": "-1Txk8iI2nJr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "kf = KFold(n_splits=3, shuffle=True, random_state = 12)\n",
        "\n",
        "cv_rid_train_r2s, cv_las_train_r2s = [], []\n",
        "cv_rid_val_r2s, cv_las_val_r2s = [], []\n",
        "\n",
        "rid_rmses_train, las_rmses_train = [], []\n",
        "rid_rmses_val, las_rmses_val = [], []\n",
        "\n",
        "X_kf, y_kf = np.array(X), np.array(y)\n",
        "\n",
        "for train_ind, val_ind in kf.split(X, y):\n",
        "    \n",
        "    X_train, y_train = X_kf[train_ind], y_kf[train_ind]\n",
        "    X_val, y_val = X_kf[val_ind], y_kf[val_ind]\n",
        "    \n",
        "    # Poly features\n",
        "    poly = PolynomialFeatures(degree=2, interaction_only=False)\n",
        "    X_train_poly = poly.fit_transform(X_train)\n",
        "    X_val_poly = poly.transform(X_val)\n",
        "    \n",
        "    # Scaling for Ridge / Lasso \n",
        "    scaler = StandardScaler(with_mean=False)\n",
        "    X_train_scaled = scaler.fit_transform(X_train_poly)\n",
        "    X_val_scaled = scaler.transform(X_val_poly)\n",
        "    \n",
        "    # Lasso\n",
        "    las = Lasso(alpha=lasso_alpha)\n",
        "    las.fit(X_train_scaled, y_train)\n",
        "    \n",
        "    cv_las_train_r2s.append(las.score(X_train_scaled, y_train))\n",
        "    cv_las_val_r2s.append(las.score(X_val_scaled, y_val))\n",
        "    las_rmses_train.append(mean_squared_error(y_train, las.predict(X_train_scaled), squared=False))\n",
        "    las_rmses_val.append(mean_squared_error(y_val, las.predict(X_val_scaled), squared=False))\n",
        "    \n",
        "    # Ridge\n",
        "    rid = Ridge(alpha=ridge_alpha)\n",
        "    rid.fit(X_train_scaled, y_train)\n",
        "    \n",
        "    cv_rid_train_r2s.append(rid.score(X_train_scaled, y_train))\n",
        "    cv_rid_val_r2s.append(rid.score(X_val_scaled, y_val))\n",
        "    rid_rmses_train.append(mean_squared_error(y_train, rid.predict(X_train_scaled), squared=False))\n",
        "    rid_rmses_val.append(mean_squared_error(y_val, rid.predict(X_val_scaled), squared=False))\n",
        "\n",
        "# Summary output\n",
        "\n",
        "print(\"Ridge\")\n",
        "print(f'Ridge mean cv R^2 (TRAIN): {np.mean(cv_rid_train_r2s):.5f} +- {np.std(cv_rid_train_r2s):.5f}')\n",
        "print(f'Ridge mean cv R^2 (VAL): {np.mean(cv_rid_val_r2s):.5f} +- {np.std(cv_rid_val_r2s):.5f}')\n",
        "print(f'Ridge mean RMSE (TRAIN): {np.mean(rid_rmses_train):.5f} +- {np.std(rid_rmses_train):.3f}')\n",
        "print(f'Ridge mean RMSE (VAL): {np.mean(rid_rmses_train):.5f} +- {np.std(rid_rmses_train):.3f}')\n",
        "print('')\n",
        "print(\"Lasso\")\n",
        "print(f'Lasso mean cv R^2 (TRAIN): {np.mean(cv_las_train_r2s):.5f} +- {np.std(cv_las_train_r2s):.5f}')\n",
        "print(f'Lasso mean cv R^2 (VAL): {np.mean(cv_las_val_r2s):.5f} +- {np.std(cv_las_val_r2s):.5f}')\n",
        "print(f'Lasso mean RMSE (VAL): {np.mean(las_rmses_train):.5f} +- {np.std(las_rmses_train):.5f}')\n",
        "print(f'Lasso mean RMSE (VAL): {np.mean(las_rmses_val):.5f} +- {np.std(las_rmses_val):.5f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qqlS88g_2wxJ",
        "outputId": "76bcce08-374a-4a1e-8217-a746dcc20caf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Ridge\n",
            "Ridge mean cv R^2 (TRAIN): 0.77624 +- 0.02028\n",
            "Ridge mean cv R^2 (VAL): 0.65596 +- 0.10234\n",
            "Ridge mean RMSE (TRAIN): 509.42303 +- 16.055\n",
            "Ridge mean RMSE (VAL): 509.42303 +- 16.055\n",
            "\n",
            "Lasso\n",
            "Lasso mean cv R^2 (TRAIN): 0.77275 +- 0.02045\n",
            "Lasso mean cv R^2 (VAL): 0.65663 +- 0.11450\n",
            "Lasso mean RMSE (VAL): 513.39005 +- 16.01409\n",
            "Lasso mean RMSE (VAL): 620.63818 +- 79.40852\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* We will keep the ridge as our by default regressor for the rest of this study"
      ],
      "metadata": {
        "id": "Uoe10nuQ4LwK"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "> **2.3 Coefficients**\n",
        "\n",
        "Let's put the light on the most explanatory variables among our variables depending on Ridge or Lasso\n",
        "\n"
      ],
      "metadata": {
        "id": "d3xWEfTZ4arn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "ridge_coefs=rid.coef_\n",
        "lasso_coefs=las.coef_\n",
        "features=sf.columns[1:]\n",
        "coefficient_data={'features':poly.get_feature_names(X.columns),\n",
        "                  'Ridge':ridge_coefs,\n",
        "                  'Lasso':lasso_coefs}\n",
        "coef_sf=pd.DataFrame(coefficient_data)\n",
        "coef_sf.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XTSJar6Y5Cdu",
        "outputId": "6570ef31-9a0e-4a60-c52f-4969ab24e338"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(136, 3)"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "coef_sf.head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "RRdr3H3e5z9A",
        "outputId": "0d5a0901-bf28-49ed-90ae-1b29f50a7609"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-bb96e6c0-29db-44a9-9e24-fab78aa4fb53\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>features</th>\n",
              "      <th>Ridge</th>\n",
              "      <th>Lasso</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>sqft</td>\n",
              "      <td>677.617878</td>\n",
              "      <td>768.802512</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>beds</td>\n",
              "      <td>24.066698</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>bath</td>\n",
              "      <td>107.870807</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>laundry</td>\n",
              "      <td>145.613574</td>\n",
              "      <td>157.711786</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>pets</td>\n",
              "      <td>-53.401127</td>\n",
              "      <td>-0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>parking</td>\n",
              "      <td>23.725012</td>\n",
              "      <td>-0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>hood_district_1</td>\n",
              "      <td>-18.363734</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>hood_district_2</td>\n",
              "      <td>-18.474859</td>\n",
              "      <td>-0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>hood_district_3</td>\n",
              "      <td>-36.690528</td>\n",
              "      <td>-4.729673</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bb96e6c0-29db-44a9-9e24-fab78aa4fb53')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-bb96e6c0-29db-44a9-9e24-fab78aa4fb53 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-bb96e6c0-29db-44a9-9e24-fab78aa4fb53');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "          features       Ridge       Lasso\n",
              "0                1    0.000000    0.000000\n",
              "1             sqft  677.617878  768.802512\n",
              "2             beds   24.066698    0.000000\n",
              "3             bath  107.870807    0.000000\n",
              "4          laundry  145.613574  157.711786\n",
              "5             pets  -53.401127   -0.000000\n",
              "6          parking   23.725012   -0.000000\n",
              "7  hood_district_1  -18.363734    0.000000\n",
              "8  hood_district_2  -18.474859   -0.000000\n",
              "9  hood_district_3  -36.690528   -4.729673"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 3. Prediction Phase using the Ridge regressor"
      ],
      "metadata": {
        "id": "qNpq_Hpp6IHi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "poly = PolynomialFeatures(degree=2, interaction_only=False)\n",
        "    \n",
        "X_train_poly = poly.fit_transform(X)\n",
        "X_test_poly = poly.transform(X_test)\n",
        "\n",
        "# Scaling for Ridge & Lasso \n",
        "scaler = StandardScaler(with_mean=False)\n",
        "\n",
        "X_train_scaled = scaler.fit_transform(X_train_poly)\n",
        "X_test_scaled = scaler.transform(X_test_poly)"
      ],
      "metadata": {
        "id": "qxcLyixsFcXN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rid = Ridge(alpha=ridge_alpha)\n",
        "rid.fit(X_train_scaled, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1oAhYHRTVypV",
        "outputId": "47e2c8f0-4125-4c05-f5ef-d020fd42b12b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Ridge(alpha=1.3226452905811614)"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Results"
      ],
      "metadata": {
        "id": "B33nYDRBV9EC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print('Ridge for training score :', rid.score(X_train_scaled,y))\n",
        "print('Ridge for testing score:', rid.score(X_test_scaled,y_test))\n",
        "print('')\n",
        "print('RMSE for training :', mean_squared_error(y, rid.predict(X_train_scaled),squared=False))\n",
        "print(\"RMSE for test: \", mean_squared_error(y_test, rid.predict(X_test_scaled), squared=False))\n",
        "print(\"\")\n",
        "print('MAE for training:',np.mean(np.abs(rid.predict(X_train_scaled)-y)))\n",
        "print('MAE for testing:',np.mean(np.abs(rid.predict(X_test_scaled)-y_test)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "miF6ivpQWDKC",
        "outputId": "381b8481-3d06-495a-80c8-f072c25b0b00"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Ridge for training score : 0.7653172379027184\n",
            "Ridge for testing score: 0.706402832216303\n",
            "\n",
            "RMSE for training : 522.7071531029383\n",
            "RMSE for test:  569.4250242124803\n",
            "\n",
            "MAE for training: 391.6989377542524\n",
            "MAE for testing: 443.68022710079765\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 4. Data visualization "
      ],
      "metadata": {
        "id": "o0gbf-R96jcG"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Comparison between actual and predicted values"
      ],
      "metadata": {
        "id": "3tshSrIFYalj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rid_predict=rid.predict(X_test_scaled)"
      ],
      "metadata": {
        "id": "byWqbiDtYhAs"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(14, 8))\n",
        "plt.scatter(rid_predict, y_test, alpha=.6) \n",
        "\n",
        "z = np.polyfit(rid_predict, y_test, 1)\n",
        "p = np.poly1d(z)\n",
        "plb.plot(rid_predict, p(rid_predict), 'm-', color='red')\n",
        "plt.plot(np.linspace(0,7000,1000), np.linspace(0,7000,1000), color='grey')\n",
        "\n",
        "plt.xlim(0, 7000)\n",
        "plt.ylim(0, 7000)\n",
        "\n",
        "plt.suptitle('Loyer Appartment à SanFransisco  (Craigslist): Actuel contre Prédite', fontsize=20, fontname='Arial')\n",
        "plt.title('Régression linéaire avec Ridge Regularization', fontsize=16, fontname='Arial')\n",
        "plt.xlabel('Valeurs prédites ($)', fontsize=16, fontname='Arial')\n",
        "plt.ylabel('Valeurs actuelles ($)', fontsize=16, fontname='Arial');\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 603
        },
        "id": "n8E8DHHlY0eU",
        "outputId": "89b82cc4-a243-492c-e3fd-7a993f664b2b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "findfont: Font family ['Arial'] not found. Falling back to DejaVu Sans.\n",
            "findfont: Font family ['Arial'] not found. Falling back to DejaVu Sans.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x576 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.savefig('ridge_actuel_predit.png')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "SX037q05Y69o",
        "outputId": "dc65b226-02b7-4f74-9557-95f5c97bb116"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 5. Using Pyspark for regression"
      ],
      "metadata": {
        "id": "6_wn7mEzIE20"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "* First let's readapt the columns of our dataset by putting the feature to predict in the end"
      ],
      "metadata": {
        "id": "Ke93MHGBMeWd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "col=sf.columns.tolist()\n",
        "col=col[1:]+col[:-len(col)+1]\n",
        "df=sf[col]\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 270
        },
        "id": "5O6PFWiYLEmU",
        "outputId": "1951ee50-7ca7-4cbb-dddf-55ad8c1b252d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-93a563d5-eec4-4f37-962d-925916af40ce\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>sqft</th>\n",
              "      <th>beds</th>\n",
              "      <th>bath</th>\n",
              "      <th>laundry</th>\n",
              "      <th>pets</th>\n",
              "      <th>parking</th>\n",
              "      <th>hood_district_1</th>\n",
              "      <th>hood_district_2</th>\n",
              "      <th>hood_district_3</th>\n",
              "      <th>hood_district_4</th>\n",
              "      <th>hood_district_5</th>\n",
              "      <th>hood_district_6</th>\n",
              "      <th>hood_district_7</th>\n",
              "      <th>hood_district_8</th>\n",
              "      <th>hood_district_9</th>\n",
              "      <th>price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3920</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>805.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3870</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5105</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4850</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1388.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.5</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4750</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-93a563d5-eec4-4f37-962d-925916af40ce')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-93a563d5-eec4-4f37-962d-925916af40ce button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-93a563d5-eec4-4f37-962d-925916af40ce');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "     sqft  beds  bath  ...  hood_district_8  hood_district_9  price\n",
              "0   805.0   1.0   1.0  ...                0                0   3920\n",
              "1   805.0   1.0   1.0  ...                0                0   3870\n",
              "2  1388.0   2.0   2.5  ...                0                0   5105\n",
              "3  1388.0   2.0   2.5  ...                0                0   4850\n",
              "4  1388.0   2.0   2.5  ...                0                0   4750\n",
              "\n",
              "[5 rows x 16 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.to_csv('sf_spark.csv', index=False)"
      ],
      "metadata": {
        "id": "f0b-jd_eM161"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* First let's import the library of pyspark"
      ],
      "metadata": {
        "id": "kWpB6CoWZ-KX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install pyspark"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8al-8VhyNAKZ",
        "outputId": "fcf84ab8-7003-4663-dacc-e7e92b981b3f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting pyspark\n",
            "  Downloading pyspark-3.2.1.tar.gz (281.4 MB)\n",
            "\u001b[K     |████████████████████████████████| 281.4 MB 44 kB/s \n",
            "\u001b[?25hCollecting py4j==0.10.9.3\n",
            "  Downloading py4j-0.10.9.3-py2.py3-none-any.whl (198 kB)\n",
            "\u001b[K     |████████████████████████████████| 198 kB 79.0 MB/s \n",
            "\u001b[?25hBuilding wheels for collected packages: pyspark\n",
            "  Building wheel for pyspark (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for pyspark: filename=pyspark-3.2.1-py2.py3-none-any.whl size=281853642 sha256=476c6f56cd8b5de9f6cad7562964f1be86a1d3ad8dc581f8937e5eab63d31912\n",
            "  Stored in directory: /root/.cache/pip/wheels/9f/f5/07/7cd8017084dce4e93e84e92efd1e1d5334db05f2e83bcef74f\n",
            "Successfully built pyspark\n",
            "Installing collected packages: py4j, pyspark\n",
            "Successfully installed py4j-0.10.9.3 pyspark-3.2.1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Let's import our dataset using **Pyspark**"
      ],
      "metadata": {
        "id": "qTpM6tsu68fZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pyspark.sql import SparkSession\n",
        "\n",
        "spark = SparkSession \\\n",
        "    .builder \\\n",
        "    .appName(\"Rent prediction in SanFransisco \") \\\n",
        "    .config(\"spark.some.config.option\", \"some-value\") \\\n",
        "    .getOrCreate()"
      ],
      "metadata": {
        "id": "x-Jv7sv7R-_G"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Load the Dataset"
      ],
      "metadata": {
        "id": "mHmkEcpO7aQw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = spark.read.format('com.databricks.spark.csv').\\\n",
        "                               options(header='true', \\\n",
        "                               inferschema='true').\\\n",
        "                               load(\"/content/sf_spark.csv\",header=True);\n",
        "\n",
        "df.show(5,True)\n",
        "df.printSchema()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JwP-nPW47dmE",
        "outputId": "9cce8fa9-8fe8-4c17-8693-7c9563c62ba0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+------+----+----+-------+----+-------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+-----+\n",
            "|  sqft|beds|bath|laundry|pets|parking|hood_district_1|hood_district_2|hood_district_3|hood_district_4|hood_district_5|hood_district_6|hood_district_7|hood_district_8|hood_district_9|price|\n",
            "+------+----+----+-------+----+-------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+-----+\n",
            "| 805.0| 1.0| 1.0|      2|   1|      2|              0|              0|              0|              0|              0|              1|              0|              0|              0| 3920|\n",
            "| 805.0| 1.0| 1.0|      2|   1|      2|              0|              0|              0|              0|              0|              1|              0|              0|              0| 3870|\n",
            "|1388.0| 2.0| 2.5|      2|   1|      2|              0|              0|              0|              0|              0|              1|              0|              0|              0| 5105|\n",
            "|1388.0| 2.0| 2.5|      2|   1|      2|              0|              0|              0|              0|              0|              1|              0|              0|              0| 4850|\n",
            "|1388.0| 2.0| 2.5|      2|   1|      2|              0|              0|              0|              0|              0|              1|              0|              0|              0| 4750|\n",
            "+------+----+----+-------+----+-------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+---------------+-----+\n",
            "only showing top 5 rows\n",
            "\n",
            "root\n",
            " |-- sqft: double (nullable = true)\n",
            " |-- beds: double (nullable = true)\n",
            " |-- bath: double (nullable = true)\n",
            " |-- laundry: integer (nullable = true)\n",
            " |-- pets: integer (nullable = true)\n",
            " |-- parking: integer (nullable = true)\n",
            " |-- hood_district_1: integer (nullable = true)\n",
            " |-- hood_district_2: integer (nullable = true)\n",
            " |-- hood_district_3: integer (nullable = true)\n",
            " |-- hood_district_4: integer (nullable = true)\n",
            " |-- hood_district_5: integer (nullable = true)\n",
            " |-- hood_district_6: integer (nullable = true)\n",
            " |-- hood_district_7: integer (nullable = true)\n",
            " |-- hood_district_8: integer (nullable = true)\n",
            " |-- hood_district_9: integer (nullable = true)\n",
            " |-- price: integer (nullable = true)\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.describe().show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sW9lSAehFtoK",
        "outputId": "8fbb0c57-9959-4309-aee9-f4d30a79399b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+-------+-----------------+------------------+------------------+------------------+------------------+------------------+-------------------+-------------------+--------------------+--------------------+-------------------+-------------------+-------------------+-------------------+-------------------+------------------+\n",
            "|summary|             sqft|              beds|              bath|           laundry|              pets|           parking|    hood_district_1|    hood_district_2|     hood_district_3|     hood_district_4|    hood_district_5|    hood_district_6|    hood_district_7|    hood_district_8|    hood_district_9|             price|\n",
            "+-------+-----------------+------------------+------------------+------------------+------------------+------------------+-------------------+-------------------+--------------------+--------------------+-------------------+-------------------+-------------------+-------------------+-------------------+------------------+\n",
            "|  count|              987|               987|               987|               987|               987|               987|                987|                987|                 987|                 987|                987|                987|                987|                987|                987|               987|\n",
            "|   mean|872.5552178318136| 1.513677811550152|1.2852077001013171|1.3333333333333333| 0.541033434650456|0.9199594731509625|0.07700101317122594|0.05977710233029382|0.020263424518743668|0.011144883485309016|0.08713272543059777|0.06281661600810537|0.08611955420466058|0.23606889564336372| 0.3262411347517731| 3441.352583586626|\n",
            "| stddev|371.5565249840777|0.9546328529434437|0.4739288471401605|0.7130579615463798|0.4985660430541476|   1.0074151036458| 0.2667282102306515|0.23719359644226862|  0.1409714611172006| 0.10503262451561672| 0.2821724359024679|0.24275583482645582| 0.2806827337962247|0.42488030448473696|0.46907439240321724|1074.0966718822808|\n",
            "|    min|              1.0|               0.0|               1.0|                 0|                 0|                 0|                  0|                  0|                   0|                   0|                  0|                  0|                  0|                  0|                  0|               225|\n",
            "|    max|           3460.0|               8.0|               4.0|                 2|                 1|                 3|                  1|                  1|                   1|                   1|                  1|                  1|                  1|                  1|                  1|              5999|\n",
            "+-------+-----------------+------------------+------------------+------------------+------------------+------------------+-------------------+-------------------+--------------------+--------------------+-------------------+-------------------+-------------------+-------------------+-------------------+------------------+\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from pyspark.sql import Row\n",
        "from pyspark.ml.linalg import Vectors\n",
        "\n",
        "def transData(data):\n",
        "    return data.rdd.map(lambda r: [Vectors.dense(r[:-1]),r[-1]]).toDF(['features','label'])"
      ],
      "metadata": {
        "id": "6MNC8iaLJUtL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Convert the data to dense vector"
      ],
      "metadata": {
        "id": "cGS80sBlNdqY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "transformed= transData(df)\n",
        "transformed.show(5)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kgGIq8NcNjLP",
        "outputId": "0ebf8ab7-3cf3-42af-f0ec-ce05959e81cd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+--------------------+-----+\n",
            "|            features|label|\n",
            "+--------------------+-----+\n",
            "|[805.0,1.0,1.0,2....| 3920|\n",
            "|[805.0,1.0,1.0,2....| 3870|\n",
            "|[1388.0,2.0,2.5,2...| 5105|\n",
            "|[1388.0,2.0,2.5,2...| 4850|\n",
            "|[1388.0,2.0,2.5,2...| 4750|\n",
            "+--------------------+-----+\n",
            "only showing top 5 rows\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Dealing with the rest of categorical varibales "
      ],
      "metadata": {
        "id": "zivyFWBaNiRH"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pyspark.ml import Pipeline\n",
        "from pyspark.ml.regression import LinearRegression\n",
        "from pyspark.ml.feature import VectorIndexer\n",
        "from pyspark.ml.evaluation import RegressionEvaluator\n",
        "\n",
        "featureIndexer = VectorIndexer(inputCol=\"features\", \\\n",
        "                               outputCol=\"indexedFeatures\",\\\n",
        "                               maxCategories=4).fit(transformed)\n",
        "\n",
        "data = featureIndexer.transform(transformed)\n",
        "data.show(5,True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XRv50Be-N6ps",
        "outputId": "085c3864-9490-4907-df6f-77254bd33cc6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+--------------------+-----+--------------------+\n",
            "|            features|label|     indexedFeatures|\n",
            "+--------------------+-----+--------------------+\n",
            "|[805.0,1.0,1.0,2....| 3920|[805.0,1.0,1.0,2....|\n",
            "|[805.0,1.0,1.0,2....| 3870|[805.0,1.0,1.0,2....|\n",
            "|[1388.0,2.0,2.5,2...| 5105|[1388.0,2.0,2.5,2...|\n",
            "|[1388.0,2.0,2.5,2...| 4850|[1388.0,2.0,2.5,2...|\n",
            "|[1388.0,2.0,2.5,2...| 4750|[1388.0,2.0,2.5,2...|\n",
            "+--------------------+-----+--------------------+\n",
            "only showing top 5 rows\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Split the data to train and test"
      ],
      "metadata": {
        "id": "v8mRUHJTOLTm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Split the data into training and test sets (40% held out for testing)\n",
        "(trainingData, testData) = data.randomSplit([0.6, 0.4])\n",
        "\n",
        "trainingData.show(5)\n",
        "testData.show(5)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f7J848OYOQcX",
        "outputId": "6cf35ee1-9d91-4fe4-a55a-53fb078c1f8a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+--------------------+-----+--------------------+\n",
            "|            features|label|     indexedFeatures|\n",
            "+--------------------+-----+--------------------+\n",
            "|[1.0,2.0,1.0,1.0,...| 3500|[1.0,2.0,1.0,1.0,...|\n",
            "|[100.0,0.0,1.0,1....| 1500|[100.0,0.0,1.0,1....|\n",
            "|[140.0,0.0,1.0,0....| 1100|[140.0,0.0,1.0,0....|\n",
            "|[225.0,0.0,1.0,1....| 1695|[225.0,0.0,1.0,1....|\n",
            "|[230.0,1.0,1.0,1....| 1100|[230.0,1.0,1.0,1....|\n",
            "+--------------------+-----+--------------------+\n",
            "only showing top 5 rows\n",
            "\n",
            "+--------------------+-----+--------------------+\n",
            "|            features|label|     indexedFeatures|\n",
            "+--------------------+-----+--------------------+\n",
            "|[200.0,0.0,1.0,1....| 1600|[200.0,0.0,1.0,1....|\n",
            "|[277.0,0.0,1.0,1....| 1800|[277.0,0.0,1.0,1....|\n",
            "|[279.0,1.0,1.0,1....| 1795|[279.0,1.0,1.0,1....|\n",
            "|[300.0,0.0,1.0,0....| 1650|[300.0,0.0,1.0,0....|\n",
            "|[300.0,0.0,1.0,0....|  225|[300.0,0.0,1.0,0....|\n",
            "+--------------------+-----+--------------------+\n",
            "only showing top 5 rows\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Import RandomForest class\n",
        "from pyspark.ml.regression import RandomForestRegressor\n",
        "\n",
        "# Calling RandomForest algorithm\n",
        "rf = RandomForestRegressor() # featuresCol=\"indexedFeatures\",numTrees=2, maxDepth=2, seed=42"
      ],
      "metadata": {
        "id": "3k5z9uIjOUMY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Using a pipeline "
      ],
      "metadata": {
        "id": "2Y1_1ju1PSzv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Chain indexer and tree in a Pipeline\n",
        "pipeline = Pipeline(stages=[featureIndexer, rf])\n",
        "model = pipeline.fit(trainingData)"
      ],
      "metadata": {
        "id": "ewArGm8tPXYk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Prediction using the randomforest regressor"
      ],
      "metadata": {
        "id": "P57uuqOQPYWo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "predictions = model.transform(testData)\n",
        "\n",
        "# Select example rows to display.\n",
        "predictions.select(\"features\",\"label\", \"prediction\").show(5)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "b2qe4DcCPeod",
        "outputId": "6a4bbd04-e572-435d-b6a2-229fd7f34300"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+--------------------+-----+------------------+\n",
            "|            features|label|        prediction|\n",
            "+--------------------+-----+------------------+\n",
            "|[200.0,0.0,1.0,1....| 1600|1783.3955575999832|\n",
            "|[277.0,0.0,1.0,1....| 1800|1995.5670682377088|\n",
            "|[279.0,1.0,1.0,1....| 1795|2252.6936234598675|\n",
            "|[300.0,0.0,1.0,0....| 1650|1838.2007492872835|\n",
            "|[300.0,0.0,1.0,0....|  225| 1994.234842362604|\n",
            "+--------------------+-----+------------------+\n",
            "only showing top 5 rows\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Evaluation part "
      ],
      "metadata": {
        "id": "xfHOQum8PvC5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Prediction and computing test error\n",
        "\n",
        "#RMSE\n",
        "evaluator = RegressionEvaluator(\n",
        "    labelCol=\"label\", predictionCol=\"prediction\", metricName=\"rmse\")\n",
        "rmse = evaluator.evaluate(predictions)\n",
        "print(\"Root Mean Squared Error (RMSE) on test data = %g\" % rmse)\n",
        "\n",
        "print(f'-'*30)\n",
        "\n",
        "#R2 score\n",
        "evaluator = RegressionEvaluator(\n",
        "    labelCol=\"label\", predictionCol=\"prediction\", metricName=\"r2\")\n",
        "r2 = evaluator.evaluate(predictions)\n",
        "print(\"R2 score (R2) on test data = %g\" % r2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w5Shgyu0P0VF",
        "outputId": "7b181d7f-b2bc-4fb3-c5e0-1181fffc689d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Root Mean Squared Error (RMSE) on test data = 663.297\n",
            "------------------------------\n",
            "R2 score (R2) on test data = 0.630182\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 6. Residual Ploting"
      ],
      "metadata": {
        "id": "yTtk9kJ-ZL-f"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Let's visualise the difference between the actual and predicted rent   "
      ],
      "metadata": {
        "id": "qA26oqRuZjqT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rid_residual=(y_test-rid_predict)"
      ],
      "metadata": {
        "id": "X0JE0mXLZPwK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(14, 8))\n",
        "plt.scatter(rid_predict, rid_residual, alpha=.7)   \n",
        "\n",
        "plt.axhline(0, linestyle='--', color='grey')\n",
        "plt.suptitle('Résidu', fontsize=20, fontname='Arial')\n",
        "plt.title('Régression linéaire avec Ridge Regularization', fontsize=16, fontname='Arial')\n",
        "plt.ylim(-2000, 2000)\n",
        "plt.xlabel('Coût prédit ($)', fontsize=16, fontname='Arial')\n",
        "plt.ylabel('Résidu ($)', fontsize=16, fontname='Arial');"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 565
        },
        "id": "IMIlfgtTZLJC",
        "outputId": "23d34b15-313d-4195-dc96-398ea09fbfae"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x576 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.savefig('ridge_residue.png')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "YCIH6hzSaSaP",
        "outputId": "2cc1c18d-6bf4-4769-c22a-e8971e521d10"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Distribution of the rental prices "
      ],
      "metadata": {
        "id": "ZfaqfSpoaWeT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(14, 8))\n",
        "ax_actual = sns.histplot(sf.price, kde=True, bins=(list(range(500, 6001, 500))))\n",
        "plt.axvline(sf.price.mean(), linestyle='--', color='grey')\n",
        "\n",
        "ax_predict = sns.histplot(rid_predict, kde=True)\n",
        "\n",
        "ax_actual.set_title('Distribution du prix des loyers à San francisco', fontname='Arial', fontsize=18)\n",
        "ax_actual.set_xlabel('Prix ($)', fontname='Arial', fontsize=16)\n",
        "ax_actual.set_ylabel('Nombre', fontname='Arial', fontsize=16)\n",
        "ax_actual.set_yticks([0, 25, 50, 75, 100, 125, 150, 175, 200, 225])\n",
        "ax_actual.set_xticks(list(range(500, 6001, 500)))\n",
        "ax_actual.set_xticklabels(list(range(500, 6001, 500)));"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 528
        },
        "id": "ZsxVatVjabub",
        "outputId": "dacafc96-5e16-4e5e-ab9b-95674a1d31e3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x576 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.savefig('Prix_distribution.png')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "FjaXr9OSayaJ",
        "outputId": "5f242b05-13d0-430c-af8c-cebea6cfff3d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    }
  ]
}